Distributed photovoltaics (DPV) can reduce electricity costs and carbon emissions in warehousing and logistics, but investment depends on load structure, capacity limits, subsidies, and surplus feed-in revenue. This study develops a Stackelberg game between a local government and an enterprise. The government sets a per-kWh subsidy; the enterprise chooses entry, capacity, and how generation is split between self-consumption and grid sales. The model incorporates the cold storage area ratio, electricity prices, installation cost, rooftop and fiscal constraints, and a fixed-mean, risk-neutral feed-in tariff benchmark. Three regimes emerge: no investment, capacity matched to internal demand, and expansion to the rooftop limit. A higher subsidy does not necessarily increase capacity. Self-consumption savings determine entry, whereas surplus feed-in revenue governs whether expansion beyond demand is profitable. A high cold storage area ratio and a sufficiently strong modeled net scale effect can still induce investment when project economics alone are insufficient. The government should offer the minimum effective subsidy at a high cold storage area ratio, and raise it to drive expansion to the rooftop limit only when marginal environmental and social benefits are high. Under surplus feed-in, a fiscal efficiency trough may arise at intermediate capacity. Numerical analyses show that the cold storage area ratio, market electricity price, feed-in tariff, and installation cost jointly determine profitability and subsidy requirements. The results support subsidies targeted by the cold storage area ratio, available capacity, and environmental benefits under budget safeguards, with enterprises sizing capacity to load, rooftop limits, costs, and surplus feed-in revenue. Managerial Relevance Statement-This study guides governments and warehousing and logistics enterprises. Government subsidy policy should move from merely raising subsidy intensity toward balancing the necessary subsidy against fiscal efficiency, with design reflecting matching between DPV generation and load, the cold storage area ratio, surplus feed-in revenue, the market electricity purchase price, and unit installation cost; subsidy caps and capacity review curb excessive subsidies. Enterprises should use the cold storage area ratio and stable demand to assess feasibility, verify the entry threshold, match capacity to demand when generation is insufficient, and expand only when surplus feed-in revenue covers incremental investment and operation and maintenance (O&M) costs. Government reviewers can apply the thresholds to screen project entry, compare demand-matched and capacities constrained by rooftop area, and reject subsidy increases that add fiscal cost without additional environmental or social benefit. Enterprises can use the same sequence to test project scale before committing capital. Sustainable Development Goals: SDG 7, SDG 9, SDG 12, and SDG 13.
Growing carbon emission reduction (CER) pressure and consumers' low-carbon awareness have driven marketplace platforms to initiate joint CER, which would theoretically lead merchants to prefer lower-cost third-party logistics (3PL) because of additional CER costs. However, some merchants actually adopt the platform's high-cost self-built logistics (SBL) after participating in joint CER. Against this background, this paper mainly explores the impact of joint CER initiated by a marketplace platform on a merchant's logistics service selection. Given that the platform takes the lead in determining CER level, we develop four models based on the merchant's decisions on participating in joint CER and logistics service selection by adopting the Stackelberg game. Counterintuitively, monotonicity analysis reveals that the merchant's profit may rise with the commission rate when consumers' low-carbon awareness is not weak. Additionally, comparative analysis interestingly indicates that when consumers have strong low-carbon awareness, even if the 3PL's cost advantage sufficiently pronounced, the merchant still selects SBL. Besides, we emphatically reveal that joint CER enhances the merchant's preference for SBL since selecting SBL can synergize with joint CER to promote demand growth. Moreover, we find that the merchant can always benefit from platform-initiated joint CER regardless of the CER investment cost coefficient. Finally, we extend our work by considering reselling mode, consumer segmentation and cap-and-trade mechanism. Through numerical studies, these extensions not only validate the robustness of core conclusions, but indicate that either more consumers with low-carbon awareness and preferences for high-value products or implementing the cap-and-trade mechanism enhances the merchant's willingness for SBL.
Deep-tier visibility is critical for supply chain networks. However, inaccurate or partial disclosure of information often leads to deviations between observed and actual deep-tier visibility. This study investigates how two types of deviations in deep-tier visibility affect the robustness of supply chain networks under disruption propagation. By utilizing grey information, we introduce two parameters, α and β, to regulate negative and positive deviations respectively. Subsequently, a disruption propagation model that incorporates these two types of visibility deviations is developed. The findings reveal a significant correlation between visibility deviations and network robustness, with the robustness being more sensitive to parameter β than to α, indicating an intriguing asymmetry in their impacts. When the probability of disruption propagation is low, enhancing the resilience of the supply chain network by controlling the visibility deviations of central firms proves more effective. Conversely, under higher disruption propagation probabilities, controlling the visibility deviations of non-central suppliers becomes more important for overall network robustness, extending the conventional focus on central firms. These results are consistently observed across three empirical supply chain networks and three synthetic scale-free networks of different sizes under the same experimental design, supporting the robustness and cross-network consistency of the findings. Furthermore, these insights offer valuable guidance for managers and policymakers aiming to improve deep-tier visibility governance and enhance supply chain network robustness against disruption propagation.
In this study, sustainable supply chain decisions are modeled under carbon tax versus cap-and-trade regulations with green credit financing. By analyzing monopoly, manufacturer competition, and demand uncertainty scenarios, we reveal a counterintuitive finding: when the intensity of manufacturers competition exceeds a certain threshold, their profits can increase with increasing green research and development costs. This is achieved through efficiency breakthroughs that reduce both emissions and compliance costs, thereby converting environmental constraints into competitive advantages. Green credit financing has threshold-dependent effects. Emission reduction occurs only when financing surpasses a critical threshold, underscoring its dual role in facilitating production and fostering innovation. Under demand uncertainty, social welfare follows a non‑monotonic path: moderate uncertainty temporarily increases welfare via retail gains, but beyond thresholds, accumulating manufacturer losses and environmental damage reduce overall welfare. Although identical carbon costs lead to uniform firm-level decisions across different regulatory frameworks, cap-and-trade systems consistently generate higher manufacturer profits. Importantly, green credit financing serves as a regulatory coordination mechanism, facilitating optimal outcomes in hybrid policy environments through dynamic quota adjustments. These findings contribute to sustainable operations by demonstrating how green credit mechanisms align profit motives with decarbonization, even under competitive and uncertain market conditions.
Consumers' inability to verify product emissions weakens manufacturers' incentives to reduce emissions, as their efforts often go unrecognized. As intermediaries between consumers and manufacturers, online retail platforms can help bridge this transparency gap through carbon disclosure initiatives. This study investigates how a platform's adoption of a carbon disclosure initiative influences the strategic participation decisions of two competing manufacturers. We show that emission transparency can backfire under joint participation, as it triggers manufacturers' more homogenous effort and pricing strategies, thereby intensifying market competition. This creates a “signaling trap” that results in a prisoner's dilemma, where both manufacturers participate in this initiative, yet both would be better off not doing so. Counterintuitively, when consumer sensitivity is high, manufacturers avoid participation to deter costly effort matching and the resulting intensified competition. Symmetric participation narrows manufacturer differentiation by inducing their heterogeneous effort and price decisions. The platform prefers asymmetric participation with only the low-emission manufacturer's participation, thus benefiting from the enlarged manufacturer differentiation and softened competition. These results indicate that carbon disclosure initiatives can either intensify or soften competition by shaping manufacturer differentiation, rather than merely serving as a disclosure mechanism. We show that participation by the low-emission manufacturer is environmentally superior to that of its high emission counterpart. Finally, we identify “win-win-win” situations where the total manufacturer profit, platform profit, and total emissions all improve under strategic platform interventions of either a service fee or a subsidy, which redirects the market away from inefficient symmetric equilibria.
Livestream e-commerce has made influencers pivotal to consumer engagement, yet influencer accidents can erode consumer trust and undermine channel performance for the collaborating brand. This paper studies how a brand can mitigate such risks by choosing between single- and multiple-influencer collaborations. We develop a hybrid-channel co-opetition model that embeds livestream accident risk and endogenous demand reallocation across channels, and use it to characterize equilibrium decisions and identify when different cooperation strategies dominate. Collaborating with more influencers can improve efficiency by reducing double marginalization, expanding demand, and softening downstream competition. However, it may also intensify intra-influencer rivalry and amplify profit losses when accident-induced demand loss becomes severe. Although influencers collectively gain reach and redundancy, higher sales do not necessarily raise their profits because internal competition compresses margins and coordination costs increase. Consumer surplus rises with higher commission rates but falls as accident risk intensifies, with the decline more pronounced under dual-influencer structures. To evaluate resilience, we propose a risk-loss index measuring the brand's profit reduction under accidents. A dual-influencer strategy improves resilience when market expansion is limited or accident risk is weak, but this advantage disappears as accident-induced demand loss intensifies. We further confirm the robustness of the main findings by extending the model to incorporate influencer information dissemination and endogenous commission rates. Our findings clarify the trade-offs among profitability, consumer surplus, and resilience in influencer collaboration design.
Global efforts toward climate change mitigation and sustainable development have increased the demand for green products and low-carbon supply chains. However, manufacturers' green investment decisions are often constrained by information asymmetry regarding consumers' green preferences. Advances in Generative AI enable third-party agencies to obtain more accurate predictive signals about market green preferences and reshape how such information is shared. This study develops a two-stage game theoretical model to examine how a third-party agency designs Generative AI-enabled information sharing strategies and how these strategies affect manufacturers' green investment decisions under heterogeneous initial carbon emission levels. We find that, in the monopolistic setting, the third-party agency does not always share information with the manufacturer. In the competitive setting, the agency either shares information exclusively with the Manufacturer-L or withholds information from both manufacturers. Specifically, the agency optimally shares information exclusively with the Manufacturer-L when the low-state green preference is sufficiently high and the initial carbon-emission gap between manufacturers is sufficiently large; otherwise, the agency refrains from sharing information with either manufacturer. More interestingly, our results reveal that competition does not necessarily increase the manufacturers' green investment levels. We further conduct numerical experiments with standardized industry data to assess the robustness of our analytical findings.
The unauthorized selling of genuine products in gray markets is a growing concern for companies worldwide. Gray markets create unfair competition by offering genuine products at different prices, which cannibalizes demand from authorized channels. Consumers often unknowingly purchase products from unauthorized channels and feel deceived when they discover that these products are unauthorized and often lack after-sales services and warranty benefits. This negatively impacts the brand image and forces the manufacturer to lower product prices to compete with the gray market, ultimately reducing the manufacturer's profit. We address this issue by analyzing how brands and manufacturers can use blockchain technology to combat gray markets. Surprisingly, we find that despite a reduction in total channel sales, blockchain imple mentation is an optimal strategy for the brand/manufacturers when both the product's brand equity and the proportion of myopic consumers in the market are relatively high. Under blockchain technology (BT) implementation, the retailer incurs losses when the proportion of myopic consumers is moderate to high, due to reduced product diversion to the gray market and a decrease in total channel sales. Additionally, we discuss the implications of blockchain adoption on consumer surplus and social welfare. Furthermore, we extend the model to explore the impact of markdown product selling, retailer's auditing policy, service competetion, uncertain demand, and variable blockchain implementation costs on adoption decisions. We find that the au thorized retailer benefits from blockchain implementation when markdown products are involved, particularly when waiting costs and channel differentiation are low. Finally, we demonstrate the robustness of our results through several practically relevant model extensions.
Rising uncertainty in renewable energy increases supply-demand fluctuations and contract fulfillment risks, making coordination between the medium- and long-term electricity (MLTE) market and the electricity spot (ES) market increasingly important. Existing studies often treat uncertainty as an exogenous parameter and ignore the dynamic execution of MLTE contracts across multiple ES periods within the delivery cycle, limiting the representation of real-world operating features in sequential electricity markets under uncertainty. This study develops a bi level game model of the MLTE and ES markets, based on a distributed stochastic optimization framework, to analyze participants' trading decisions and market equilibrium. The upper-level model uses Nash bargaining to determine contracts in the MLTE market, and maps contracted quantities to the ES market through a decomposition curve. The lower-level model describes participants' bidding decisions under a uniform clearing price mechanism based on the decomposed quantities and execution prices. To capture uncertainty, a higher-order Markov model is used to represent dynamic scenario transitions, and a distributed optimization algorithm is applied to solve the model. The results show that, firstly, MLTE contracts can effectively reduce ES price volatility and stabilize participants' profits. Secondly, under uncertainty, financial contracts are more effective than physical contracts in promoting renewable energy consumption. Finally, although MLTE contracts can improve the bargaining power of renewable generators in the ES market, they do not guarantee a dominant share of traded quantities. Overall, this study provides reliable quantitative evidence and managerial insights for trading decisions and the design of sequential electricity markets under uncertainty.
Can a national technology strategy, designed to accelerate industrial upgrading, inadvertently erode the gender diversity of the R&D workforce it seeks to empower? This study investigates this critical trade-off in engineering and technology management by examining China's “Made in China 2025” (MIC2025) policy. Integrating Tokenism Theory with the literature on state-directed innovation tournaments, we theorize how macro-level policy pressures activate and amplify micro-level gender barriers within R&D teams. Employing a rigorous staggered difference-in-difference-in-differences (DDD) design reinforced by stringent multi-way interactive fixed effects and comprehensive identification diagnostics, and a novel bidirectional long short-term memory (Bi-LSTM) attention gender predictor applied to over 20 million patent inventors, we find that MIC2025, while boosting innovation output, reduced the female inventor share in treated firms by 5 percentage points, a 25% decline relative to the sample mean. Absolute count analyses confirm this represents a genuine contraction in female innovative output, not merely relative dilution. The policy's adverse effect operates primarily through a clustering of authorship credit away from women, with suggestive evidence of a secondary job-demands channel via intensified overtime. The negative impact is concentrated in incremental utility model patents rather than frontier inventions, and is significantly attenuated by female chief technology officers (CTOs) but amplified by female executives in non-technical roles, a divergence we explain through Role Congruity Theory. These findings reveal a tension between technology upgrading and workforce inclusion, and identify the functional positioning of female technical leadership as a key lever for sustaining R&D diversity under competitive pressure.
How should supply chain partners design and govern blockchain-enabled anti-counterfeiting when consumer psychology alters demand incentives? We address this question in a supply chain where deceptive counterfeits create authenticity uncertainty and consumers may anticipate regret after choosing an authentici ty-uncertain purchase over a more reliable alternative. We develop Stackelberg game models under two blockchain implementation leadership structures: a manufacturer-led scenario (Scenario MB) and a retailer-led scenario (Scenario RB). We derive the equilibrium blockchain-enabled verification effort and pricing decisions under both structures. Key findings include: First, a higher block chain-enabled verification effort is not always better. When authentic product quality is low, a moderate verification effort is optimal because the incremental benefit from stronger verification may not offset the additional cost, whereas when authentic product quality is sufficiently high, maximizing verification effort becomes optimal. Second, the retailer generally prefers manufacturer-led implementation due to free-riding. However, interestingly, this preference is not always shared by the manufacturer. Specifically, the manufacturer prefers to lead blockchain implementation and bear the associated verification cost when authentic product quali ty is low. Third, while conventional wisdom suggests that CAR reduces consumer willingness to pay, which is detrimental to supply chain stakeholders, we reveal a counterintuitive outcome: under certain conditions, CAR can actually benefit the stakeholder that does not lead blockchain implementation. Practice based nu merical analysis shows that residual counterfeit infiltration is reduced by 73.58% and 29.95% under manufacturer and retailer leadership, respectively, relative to conventional verification. We further examine four extensions to assess robustness.
Amidst the emphasis on sustainability, refurbishing market has seen widespread development. Refurbished products (RPs) can directly compete with new products (NPs) as discounted alternatives. However, empirical research shows that consumers have valuation uncertainty for RPs because of the opaque refurbishing process. When NPs and RPs coexist, such uncertainty may lead to inaccurate utility comparisons, which triggers anticipated regret. In the above context, we study competitive pricing decisions of a new-product platform and a refurbished-product platform. Considering that the quality of RPs depends on that of NPs, quality can serve as another lever for the new-product platform. Hence, we further investigate its quality decision. We find that: When quality is exogenous, varying quality levels lead to different pricing interactions: Low level prompts the refurbished-product platform to adopt a relatively lower price compared to that of NPs. At this point, the refurbished-product platform benefits from the increasing sensitivity to anticipated regret, while the new-product platform experiences the opposite effect. Conversely, high level causes the price of RPs to gradually approach that of NPs. At this point, the anticipated regret affects these two platforms in the same direction. When quality is endogenous, the optimal quality level for the new-product platform is at a compromise, however, it is always lower than that under the monopoly, and the gap is exacerbated by the increasing sensitivity to anticipated regret. Despite the above, the consumer surplus is still better than that under the monopoly, and this advantage becomes more pronounced when the proportion of high-valuation consumers increases.
Generative artificial intelligence (GenAI) is rapidly transforming digital platforms. Unlike earlier AI technologies centered on prediction and classification, GenAI's creational capabilities, probabilistic outputs, and iterative learning dynamics reshape how value is created, captured, and governed within platform ecosystems. This editorial synthesizes insights from the Special Issue (SI) on “The Transformative Power of Generative AI for Digital Platforms,” which received 38 submissions and published 6 articles spanning organizational, technological, strategic, behavioral, and societal perspectives. Drawing on the existing literature, the editorial develops a multidimensional framework that organizes the dimensions of GenAI-driven platform transformation: adoption and organizational embedding; creational affordances and innovation mechanisms; strategic and operational reconfiguration; human and behavioral responses; and governance, risk, and legitimacy. We further conceptualize the dimensions as a recursive generativity-governance cycle linking embedding, affordance actualization, platform reconfiguration, stakeholder responses, and governance feedback. The published articles are subsequently mapped onto this framework, highlighting their contributions to both literature and practice. The editorial concludes by outlining theoretical implications, managerial considerations, and directions for future research, offering a comprehensive understanding of GenAI as a socio-technical force reconfiguring platform architectures, decision processes, human-AI interaction, and institutional environments.
This editorial introduces the Special Issue “Operational Innovation in Interdisciplinary Research” and examines how operational innovation advances engineering management (EM) under operational risks and sustainability pressures. Building on 18 accepted papers selected through a rigorous peer-review process, the editorial frames operational innovation as the development and application of novel technologies, policies, and analytical methods that improve operational outcomes. The contributions are organized across six interdisciplinary interfaces including EM with marketing, finance, logistics, procurement, manufacturing, and retailing. Together, the papers show that technological, institutional, and methodological innovations can strengthen supply chain resilience and support environmental, social, and governance objectives, but their effectiveness depends on incentive alignment, capability fit, market structure, and stakeholder behavior. The editorial concludes by highlighting operational innovation as a fertile interdisciplinary research agenda for scholars, practitioners, and policymakers pursuing sustainable and resilient operations. It also identifies future opportunities for cross-domain theory building and evidence-based governance.
Despite the growing interest in how interorganization relationships shape sustainable behaviors of firms, the effects of relationships among supply chain parties on firms' environmental, social, and governance (ESG) performance remain underexplored. Drawing on practice diffusion theory, this study examines how the ESG performance of suppliers and that of customers affect the ESG performance of the focal firm; additionally, the moderating roles of supply chain concentration and overlapping ownership are also explored. Employing a panel dataset of Chinese A‑share list firms from 2009 to 2023, we find that both suppliers' and customers' ESG performance positively influence the focal firm's ESG performance. After a series of robustness and endogeneity tests, these findings remain robust. Notably, the diffusion effect intensifies when the focal firm maintains concentrated relationships or has overlapping ownership with its supply chain partners. Conversely, unconcentrated relationships and the absence of overlapping ownership attenuate this diffusion effect. This study emphasizes the importance of relationship type and relationship intensity in understanding ESG diffusion in a supply chain and offers insights for research on practice diffusion and sustainable supply chains.
The gaming industry is thriving, propelled by rapid advances in digital technology and a steadily expanding player base. Game publishers adopt a variety of release strategies: Some games are released on schedule, while others experience delays, a phenomenon commonly referred to as “slips.” Moreover, the growing size of the gaming community also generates significant network effects: as more players engage with a game, enthusiasm spreads, enriching the overall player experience. This paper examines the release strategies of game publishers in a competitive market where player engagement is shaped by network effects. Utilizing a game-theoretic model, we find that (1) immediate release is optimal when the initial quality gap is small or the cost of delay is high, (2) a fast slip strategy dominates when the quality gap is moderate and can be substantially reduced in a short time, and (3) a slow slip strategy is preferred when the gap is large, quality can be improved rapidly, and postponement is relatively inexpensive. The paper also considers the possibility of “reverse slip”, releasing earlier than planned. Extending the model to competing publishers, we derive equilibrium conditions for release-timing choices and show that both slip strategies remain viable, depending on quality differences among games. We further substantiate these analytical results with real-world numerical evidence from major game releases, showing that well timed acceleration or postponement can improve launch outcomes, whereas releasing with unresolved quality problems may lead to severe post-launch penalties. Our findings offer actionable guidance for release planning that jointly considers quality evolution, network effects, and competitive dynamics.
To sustain competitive advantage, firms increasingly adopt open innovation (OI) by leveraging external collaborations. While prior research highlights the importance of supply chain relationships for innovation, how supply chain dependency shapes product-related OI remains underexplored. Integrating Resource Dependence Theory (RDT) and the Knowledge-Based View (KBV) through the key knowledge dependency, this study examines how supplier concentration (SC) and customer concentration (CC) differ in their relationships with OI, and how main business uniqueness (MBU) and betweenness centrality (BC) moderate these relationships. Using panel data from Chinese listed manufacturing firms from 2013 to 2022, we empirically test these relationships. To ensure the research rigor, we subject the findings to comprehensive robustness checks (alternative measures, lagged designs, Heckman correction for sample selection, and 2SLS estimation for endogeneity) and further validate them through K-means cluster analysis. The results reveal some key insights: (1) SC exhibits a significant negative relationship with OI, whereas CC shows a positive association; (2) MBU strengthens the link between CC and OI, while high BC amplifies SC's negative effect and attenuates CC's positive effect; and (3) SC's negative association with OI consistently dominates CC's positive association across configurations. Moreover, MBU's moderating role is significant only in low SC-high CC clusters and BC exhibits varying effects depending on the supply chain configuration. These findings highlight the critical role of dependency structure and position in innovation and provide configurational guidelines for managing OI in complex supply chain contexts.
Live streaming has become increasingly popular, prompting sellers to select influencers from Multi-Channel Networks (MCNs) to attract consumers. However, MCNs may exaggerate influencers' historical sales data, creating reliability asymmetry that can mislead sellers about demand-enhancing capabilities and potentially result in losses. To address this, an MCN can reveal its reliability through negotiation with the seller (Strategy CN), adopt blockchain technology (Strategy BT), or choose non-disclosure (Strategy NI). Despite the prevalence of these strategies, their impact on sellers' decisions in live streaming remains underexplored. Our analysis constructs three models based on MCNs' strategies, focusing on how sellers decide on influencer selection and pricing. First, we find that a higher commission may create a double-loss trap, harming both the seller and the MCN. Intuitively, negotiation should lower prices and blockchain adoption should raise them; however, our analysis reveals the opposite when seller bargaining power and consumer BCT sensitivity are both low. Second, Strategy CN consistently promotes live-streaming adoption and Strategy BT may discourage participation when consumer attention to live streaming is low. Third, low-reliability MCNs may also benefit from information sharing. Finally, we conduct comprehensive robustness checks through four extensions. In particular, under PRI-based information acquisition, we identify asymmetric effects of seller-observed signals across MCN types, which provide managerial insights for designing information disclosure strategies in live-streaming.
